Multi-View Reconstruction with Global Context for 3D Anomaly Detection

2025/07/29 by Sun, Yihan, Cheng, Yuqi, Cao, Yunkang +2
#Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2507.21555

Abstract

3D anomaly detection is critical in industrial quality inspection. While existing methods achieve notable progress, their performance degrades in high-precision 3D anomaly detection due to insufficient global information. To address this, we propose Multi-View Reconstruction (MVR), a method that losslessly converts high-resolution point clouds into multi-view images and employs a reconstruction-based anomaly detection framework to enhance global information learning. Extensive experiments demonstrate the effectiveness of MVR, achieving 89.6% object-wise AU-ROC and 95.7% point-wise AU-ROC on the Real3D-AD benchmark.

Citations

• A Comprehensive Survey for Real-World Industrial Defect Detection: Challenges, Approaches, and Prospects
• Towards High-Resolution 3D Anomaly Detection: A Scalable Dataset and Real-Time Framework for Subtle Industrial Defects
• Boosting Global-Local Feature Matching via Anomaly Synthesis for Multi-Class Point Cloud Anomaly Detection
• Exploring Intrinsic Normal Prototypes within a Single Image for Universal Anomaly Detection
• R3D-AD: Reconstruction via Diffusion for 3D Anomaly Detection
• Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detection
• PointCore: Efficient Unsupervised Point Cloud Anomaly Detector Using Local-Global Features
• Towards Scalable 3D Anomaly Detection and Localization: A Benchmark via 3D Anomaly Synthesis and A Self-Supervised Learning Network
• Vision Transformers Need Registers
• Real3D-AD: A Dataset of Point Cloud Anomaly Detection
• ReContrast: Domain-Specific Anomaly Detection via Contrastive Reconstruction
• Stable and low-precision training for large-scale vision-language models
• Multimodal Industrial Anomaly Detection via Hybrid Fusion
• On the Convergence of Adam and Beyond
• Open3D: A Modern Library for 3D Data Processing

Related

• SiM3D: Single-instance Multiview Multimodal and Multisetup 3D Anomaly Detection Benchmark
• MC3D-AD: A Unified Geometry-aware Reconstruction Model for Multi-category 3D Anomaly Detection
• 3DKeyAD: High-Resolution 3D Point Cloud Anomaly Detection via Keypoint-Guided Point Clustering
• Multimodal Industrial Anomaly Detection via Hybrid Fusion
• Towards Zero-shot Point Cloud Anomaly Detection: A Multi-View Projection Framework